I have 5x WD20EARS configured in RAID 5 in a Sans Digital TowerRAID TR5UT-B (hardware RAID) with the included Highpoint Tech R622 Controller Card. The unit is hooked up to a Windows 2008 box with eSATA and is accessed over a network share by no more than two people. The drives are primarily being used for archival purposes of photoshop and indesign files. They don't have to be fast, just large and dependable enough to not lose data.
We've copied about a TB of data to the RAID when drive 0 of the array failed. I'm doing an RMA with Western Digital, but they tell be this drive with "Advanced Format" is not supported in RAID configurations. When I got the drives I just poped them into the enclosure and formated everything with defaults within Windows. Is there anything I can do to help the situation and avoid future failures without having to get all new drives? I've read that TLER may play a factor but that these not drives may not even support changes made by the utility.
TLER is what it's all about. WD won't support it's non-RAID drives in RAID systems because the RAID controller may falsely declare the drive "dead" when it's really just busy doing error recovery. So simply having a RAID controller declare the drive as failed isn't sufficient cause for WD to accept that the drive is actually dead.
You can certainly use non-TLER drives in a RAID array, but don't expect to get in-warranty replacements unless you can prove the drive is non-functional even when used as an independent, non-RAID drive.
I'm not sure if your main concern is getting the failed drive replaced or if you're worried that other drives may also be declared dead by the RAID controller.
If it's the former, then you could try running drive tests with the drive as an individual unit not part of a RAID array. If you can find a problem with it then you can return it to WD and tell them that it's dead and the problem has nothing to do with RAID. If you can't find a problem with it then you could try putting it back into the array to see if it will work OK there. It's quite possible that the drive has marked one or more marginal sectors that caused the original problem as bad and it won't use them again.
If it's the latter, then yes, I think you're probably SOL. But the chances of a non-TLER drive being dropped by the RAID array due to error recovery timeouts are probably relatively uncommon - rather than replacing all the drives immediately you might consider just running the ones you have and not replace them until/if they're declared dead by the RAID controller. There's a fair chance that won't happen until a drive REALLY dies due to more serious problems. And if you can verify the problems independent of a RAID array then you'll be able to get warranty replacement if the failure happens within the warranty period.
Firmware upgrades are often available for raid controllers that don't currently support 4k blocks on the newer drives such as the WD advanced format drives. I would check to see if you hardware currently support this or if there is improved firmware.
Secondly, I would check the drive itself. Hook it up directly to a pc and run WD's free diagnostic tools.
If the raid supports 4k drives, and the drive passes WD tests then I would conclude you have a TLER problem and it would be in your beast interest to buy 24/7 raid rated drives like WD RE series or seagate ES